Second Brain
Set up and operate a file-based second brain — a knowledge vault that every Claude Code session reads on start and writes back to before it ends, so context compounds across sessions instead of evaporating. Use "second-brain setup" to bootstrap a new vault (folder structure, brain CLAUDE.md, note templates, optional session-log hook), and follow the OPERATE rules in every session once a vault exi…
Review Panel
Use when any substantive work product has been generated — copy, emails, documents, proposals, plans, code, designs, skills, newsletters — and is about to be delivered, shipped, marked done, or given a quality verdict. Also use when the user asks for a review or critique of existing work. Trigger BEFORE declaring anything ready or presenting it as finished.
Vault Gardener
Recurring maintenance pass for a second-brain vault — link hygiene, naming, metadata/tags, Inbox filing, dedupe, learnings-index sync, index-note freshness. Run daily as a scheduled agent (or weekly by hand). Keeps recall precise as the vault grows; without gardening, a knowledge vault decays into a junk drawer with a search bar. Use when asked to "garden the vault", "clean up the vault", or as t…
Storm Research
Use when researching any topic, question, market, company, technology, or decision — before writing reports or proposals, making business or investment decisions, preparing for negotiations, interviews, or presentations, or learning a new field. This is the default research method for all research tasks. Skip only for single-fact lookups, or when the user explicitly asks for a quick answer or nam…
Extract Approach
Reasoning recorder — fires automatically after cracking a HARD problem (multiple failed attempts, non-obvious insight, judgment call between viable options, or an invented quality bar). Captures HOW the problem was solved — problem shape, dead ends, the turning point, the transferable recipe — as one note in your learnings directory, so the reasoning survives the session and any model change. Do…
Focus Group
Synthetic audience pretesting (Justin Brooke "predictive wear" method). Run ANY outbound content — ads, emails, sales pages, newsletters, social posts, offers — through a per-business panel of AI persona dossiers BEFORE it ships. Panel critiques → copywriter rewrites 3 variants → prediction engine scores 0-100 → iterate until 85+. Intended as a hard gate: no outbound content ships without a run.…